--- MITgcm/verification/global_with_exf/README 2003/08/07 02:31:29 1.3 +++ MITgcm/verification/global_with_exf/README 2003/08/15 01:42:44 1.4 @@ -1,4 +1,4 @@ -heimbach@mit.edu 07-Nov-2002. + heimbach@mit.edu 07-Nov-2002. This verification experiment is almost identical to the global_ocean.90x40x15 experiment, except that it uses the exf (and cal) package instead of the @@ -18,19 +18,23 @@ menemenlis@jpl.nasa.gov 05-Aug-2003 -On-the-fly spatial interpolation has been added to pkg/exf. +Input-field spatial interpolation has been added to pkg/exf. It is enabled by defining CPP option USE_EXF_INTERPOLATION in EXF_CPPOPTIONS.h or in ECCO_CPPOPTIONS.h -Both bi-linear and bi-cubic (#define BICUBIC) are supported. +Both bi-linear and bi-cubic interpolation schemes are supported. -This package is a placeholder until more general coupler -is made available by ESMF project. It assumes that input -grid is Cartesian with arbitrary latitude increments in the -y-direction but with equidistant longitude increments in +This package is a placeholder until a more general coupler +is made available by the ESMF project. The output grid can be +arbitrary (cubed-sphere should be OK), but it is assumed that +the input grid is Cartesian with arbitrary latitude increments +in the y-direction and with equidistant longitude increments in the x-direction. The input grid must encompass the complete output grid in the y-direction (i.e., extrapolation is not -supported) and it is assumed that the grid is periodic in -the x-direction. +supported). It is also assumed that the grid is periodic in +the x-direction; to use the interpolation routine with a +non-periodic domain, make sure there is sufficient padding +at the edges, i.e., two points for bicubic and one for bilinear +interpolation. The verification/global_with_exf experiment has been modified to use the USE_EXF_INTERPOLATION option, the input grids @@ -43,5 +47,5 @@ *_nlon and *_nlat :: input x-grid and y-grid size In this particular example the input and output grids are -the same, so the results of the verification experiment remained +the same, so the results of the verification experiment remain unchanged from before.